We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Meiirbek Islamov - Rapid prediction of Thermal Transport in Metal-Organic Frameworks | SciPy 2023
Learn how machine learning and molecular dynamics simulations can enable rapid prediction of thermal transport in metal-organic frameworks, potentially revolutionizing gas adsorption applications.
- Rapid prediction of thermal transport in metal-organic frameworks is important due to their potential in gas adsorption applications
- 90,000 synthesized MOFs and 500,000 predicted MOFs exist, but only a small fraction have experimentally measured thermal conductivities
- Machine learning and molecular dynamics simulations can be used to predict thermal conductivities
- Graph neural networks are a good choice for training models due to their ability to learn complex relationships between structure and property
- The topology of MOFs is important for heat transport and can be used to predict thermal conductivities
- Defects in MOFs can also improve thermal conductivities, especially correlated defects
- Thermal conductivity is reduced in certain directions due to perpendicular interactions
- Larger pore sizes can lead to higher thermal conductivities
- MoF design space is huge, covering 97% of all terminal connected data
- Hypothetical MOFs can be created and studied to learn more about thermal conductivities
- Model prediction is important in identifying direction-dependent thermal conductivity
- The thermal conductivity of a MOF is influenced by its topology, structural, and compositional properties.